Metrics

Every service on ZevCloud reports what it is doing while it runs. You can see live usage and recent trends without leaving the dashboard, so you can tell whether a service is healthy, why it might be slow, and whether you are getting close to your plan limits.

Where to find metrics

There are two views:

• Per service. Open any service and go to the Metrics tab. This shows that one service in detail, with charts.
• Across your team. The Metrics page in the sidebar shows every service in your team at a glance, one card each. Click a card to jump straight to that service’s Metrics tab.

Time range

At the top of the metrics view you can switch between 1H, 24H, and 7D. The live tiles always show current values. The charts follow the range you pick, so you can look at the last hour for a spike or the last week for a trend.

What each number means

Metric	What it shows
CPU	Processor use as a share of the CPU your plan includes, so the percentage is against your own allowance (it can read above 100% on a multi-core plan).
Memory	RAM in use versus your plan’s memory limit. Sitting near the limit for long is a sign to optimize or move up a plan.
Network	Traffic in and out of your service, shown as a live rate over the selected range.
Disk I/O	How fast your service is reading and writing to disk. This is activity, not capacity.
Disk usage	How much storage your service’s own files take up. This is separate from Object Storage buckets.
Connections	For a managed database: active client connections against the maximum your plan allows.
Build	For apps that build from a repository: the size of the built image and how long recent builds took.

Disk usage is not your storage buckets

“Disk usage” on a service is the space that one service’s own files use. Your Object Storage buckets are a separate product with their own usage, shown under Storage. A database’s disk usage and a bucket’s size are two different numbers.

Databases: connection count

A managed database adds a Connections tile showing active connections against your plan’s limit. Watching it helps you catch a connection leak (an app that opens connections and never closes them) before it exhausts the pool and new requests start failing. If this number climbs and stays high while traffic is normal, look at how your app manages its database connections.

Applications: build size and time

An application that builds from a repository shows a Build section with the size of its built image and the duration of recent builds. If a change makes your build noticeably slower or your image larger, you will see it here.

Give new metrics a minute

Metrics start appearing a couple of minutes after a service is running, and the trend charts fill in over the first hour. A service that is stopped or still deploying does not report live metrics.

Who can see metrics

Anyone on your team who can view a service can see its metrics. There is nothing to switch on. Metrics are available by default for every running service, on every plan.